This year we are honoured to have Professor Paul Harrigan, Deputy Dean and Professor of Marketing in UWA Business School at the University of Western Australia, present on the topic
"Truth, social. Where are we in the fight against misinformation?"
Abstract:
The rapid spread of misinformation on social media, amplified by generative artificial intelligence, poses a significant global threat. In response, this study develops an explainable AI-based decision support framework for misinformation detection using transformer-based language models. Leveraging a large real-world dataset from Meta, we benchmark encoder-only, encoder-decoder, and decoder-only models in both general-purpose and domain-adapted forms. Results show that domain-specific fine-tuning significantly boosts performance, with a fine-tuned GPT model achieving the highest prediction accuracy, outperforming all other architectures. Importantly, we introduce two explanation strategies—one combining our best fine-tuned encoder-decoder model, BART with LIME, and GPT for feature-based reasoning, and another using GPT-4o-mini for end-to-end prediction and explanation—both designed to improve the interpretability and trustworthiness of artificial intelligence decisions for content moderators.
About Professor Paul Harrigan
Professor Paul Harrigan is Deputy Dean and Professor of Marketing in UWA Business School at The University of Western Australia (UWA). He is also an Adjunct Professor at IESEG School of Management in France, and Editor-in-Chief of the Australasian Marketing Journal.
Paul has a PhD from Ulster University in the UK (2008), was previously a Lecturer in Marketing at The University of Southampton (2008-12), and has been at UWA since 2012. Paul was also Vice-President of the Australasian and New Zealand Marketing Academy (ANZMAC) 2019-2024.
Paul's research expertise runs across marketing and information systems, specifically digital transformation and misinformation. He has published his research in over 60 international journal articles and books, and presented it at over 60 international conferences.
Current projects investigate the use of debunking and prebunking methods in combatting misinformation on social media, the impact of slow motion video on social media follower engagement, and how to reduce the harm of social media ads on users’ body image.
This year, the distinguished Oration event will be held
WHERE
UWA University Club
Entrance 1, Hackett Drive
Crawley WA 6009
WHEN
Wednesday 15 October
Pre-dinner drinks from:
18:15PM AWST
Event Start:
7:00PM AWST
EVENT PRICE
Member Concessional:
$115.00
Member: $160.00
Non Member: $180.00
**Dress Code: Formal dress attire**
The winners of the 1962 Prize and 1962 Medal will also be announced. These prestigious awards, sponsored by Professor Dennis Moore AM, showcase the best and brightest minds in WA and are a celebration of local talent, student excellence and the next generation of ICT Professionals.
ADDITIONAL BOOKING INFORMATION:
- Event Terms & Conditions can be found on the registration page.
- To confirm attendance and CPD point allocation, on arrival at registration, please make sure to have your event confirmation e-mail with QR code ready to present where your attendance will be recorded. This can be provided via your phone or a printed copy.
- If you are not a current ACS member, you will be asked to 'create an account'. Your details will be used for registration processes only. If you experience technical issues, please contact the WA Branch by e-mail acs.wa@acs.org.au
Annual Dennis Moore Oration - Past Orators
Since 2012, to commemorate fifty years of digital computing in Western Australia, the WA Branch of the ACS has invited a distinguished scholar and researcher with a connection to WA to present a lecture on the leading edge of an important and emerging area of information and computer technology.
Press the play button to watch the short video of past WA Dennis Moore Orators and the topics presented
For the Best Experience, Rotate Your Device
For an optimal viewing experience on a mobile device, please rotate your phone to landscape mode.